Triangulation Issues

[francisco_bobadilla]

Does anyone knows how to fix this triangulations.  Adding extra breaklines does not seem to fix the issue.  Thanks in Advance

[mfsamee]

Hi,

Please elaborate your question.

[Jeff_M]

Set your Use Max Length triangle to On and give it a value shorter than those unwanted lines.

[francisco_bobadilla]

Hmmm.  Didn't seem to fix the issue :(.  Still get these lines, I even played with the triangle angle setting

[dasindog]

maybe post a drawing?

[fcernst]

Start Flippin' Faces...guess it's called Swap Edges now.

[francisco_bobadilla]

Too large to post.  I can email it to you.

[sboon]

How was this TIN created - maybe a corridor, a grading infill or something else?

Steve
Please use the Accept as Solution or Kudo buttons when appropriate

[francisco_bobadilla]

The TIN was created using different definition methods: pasting corridor surfaces and by  feature lines.

[Jeff_M]

Feel free to email it me, Francisco.

jeffm AT quuxsoft DOT com

[troma]

I fail to see what is 'wrong' with the TIN.  Sure it looks a bit funky, but that's only natural when you have one point on the left that is triangulating to a hundred on the right.  Maybe if you'd used a larger mid-ordinate distance on the curved breakline it would look a bit nicer.  To me it looks very sparse for data; I feel you need to add information in the middle somewhere to tell the surface what to do.  That applies just as much to the 'good' area of your surface as it does to the 'bad' area.

Or perhaps you can elaborate on what is 'wrong' with the triangulation, so I can understand what you're aiming for?

[francisco_bobadilla]

You can see the contours get tighter at some spots.  This is not correct.  I will try the mid-ordinate distance setting

[SurveyorLee]

Looks like there are points that are not in the surface and no fault lines. 

Surveyor Lee

[francisco_bobadilla]

What do you mean by fault lines?

[fcernst]

Yes, start Swapping Edges...

The more experienced and accomplished you become at reading and interpreting the triangulation of TIN surfaces, you will find that you will have the ability to create surfaces solely from point data.

From the C3D Help:

"...Given a set of data points, Delaunay triangulation produces a set of lines connecting each point to its natural neighbors."

It's true! For the most part you can create roads, channels, ditches and even walls solely from points.The surface algorithm in Civil 3D will create the surface per the algorithm . The onus is on the author of the surface to ensure the surface is a correct interpretation per the given data. I'm not advising abandoning the use of breaklines, it is good practice to use Breaklines, however the more experienced you become, you will develop the skill set to be able to quickly visually scan the triangulation of a surface, and note perhaps a few areas where the triangulation needs to be adjusted.

Experienced surveyors have to do this all the time. Before they deliver their final product to us, which is the Existing Surface, they have to finalize the surface  triangulation per how they observed and interpreted the surface in the field, and how they want to portray and deliver that finalized surface as their product to the world. Sometimes, we see mistakes in the triangulation of these final existing surface products, that need to be corrected, such as areas at culvert headwalls, ditch flowlines and banks, retaining walls.

So now with this information and background, you can edit and correct your surface as you see fit by adjusting the triangulation.

You say this is mish mash of pasted surfaces, etc.... You may want to review that definition process on this surface, just to see if it really is making sense and good progress towards you final goal.